Objectives for Responsible Development of AI System
Set clear goals for developing AI responsibly and integrate them into its development lifecycle.
Plain language
This control is about setting clear goals to develop AI systems responsibly. Imagine your customer service AI gives wrong answers because it was not built with proper guidelines. By setting objectives, you help ensure your AI behaves correctly and doesn't disappoint or harm your customers.
Framework
ISO/IEC 42001:2023
Control effect
Preventative
Classifications
N/A
Official last update
01 Dec 2023
Control Stack last updated
19 May 2026
Maturity levels
N/A
Official control statement
The organisation shall identify and document objectives to guide the responsible development AI systems, and take those objectives into account and integrate measures to achieve them in the development life cycle.
Why it matters
If AI objectives aren't set, the AI might mislead customers, causing loss of trust or legal issues. Imagine an AI advisor giving wrong financial advice to users, risking financial harm.
Operational notes
Review and update AI objectives whenever changing the system's purpose or adding significant features, not just annually.
Implementation tips
- The AI lead should document specific goals for the AI system, like ensuring it provides accurate answers. A simple list with examples of correct and incorrect AI responses could illustrate these objectives well.
- The product owner should integrate these AI objectives into the product design. During team meetings, discuss how new features align with responsible AI goals, perhaps using a simple checklist.
- Procurement should ensure vendors are aware of these AI objectives when supplying AI tools. Add a clause in purchase agreements requiring vendors to adhere to these goals, ensuring a shared responsibility.
- The data steward should create processes for evaluating AI outcomes against the set objectives. Implementing basic user feedback collection to detect when the AI underperforms can help adjust objectives effectively.
- The head of risk should evaluate potential risks of not meeting these objectives and document them in a risk register. Highlight possible customer impacts, like the AI giving bad advice, and plan mitigations.
Audit / evidence tips
- AskRequest the list of objectives for the AI system. GoodThe objectives are documented, clear, specific, and aligned with the AI's intended use.
- AskAsk for the product development plan that includes AI objectives. GoodThe plan explicitly integrates AI objectives at key development stages.
- AskSee vendor contracts for AI systems. GoodContracts include AI objectives, ensuring vendor accountability.
- AskRequest recent risk assessments. GoodThe assessment identifies risks related to AI objectives and suggests mitigations.
- AskRequest user feedback reports. GoodFeedback is analysed for alignment with AI objectives, and corrective actions are taken.
Cross-framework mappings
How Annex A 6.1.2 relates to controls across ISO/IEC 27001, ISO/IEC 42001, Essential Eight, and ASD ISM.
ASD ISM
| Control | Notes | Details |
|---|---|---|
| handshake Supports (3) expand_less | ||
| ISM-0041 | Annex A 6.1.2 requires the organisation to identify and document responsible AI development objectives and integrate measures to achieve ... | |
| ISM-1998 | Annex A 6.1.2 requires defined responsible AI development objectives and their integration into development activities | |
| ISM-1999 | Annex A 6.1.2 requires the organisation to identify, document, and integrate objectives for responsible AI development into the AI develo... | |
These mappings show relationships between controls across frameworks. They do not imply full equivalence or certification.
Want to implement this AI control?
Mindset Cyber runs PECB-accredited ISO/IEC 42001 training that maps directly to the AI controls in this library.